News & Releases

What's new in Harvestry

Release notes, feature announcements, and updates from Archetyp Mobility.

v1.3.5 October 2026

Version 1.3.5: Large Turbo transcribes about five times faster — and Harvestry is far better tested.

Someone trying Harvestry wrote in recently: each 90-minute lecture took over half an hour to transcribe with Large Turbo on a MacBook Air. Was it even using the GPU? The heavy lifting had been happening all along on the Neural Engine, the part of Apple silicon built for exactly this. The real problem was ours: what we'd shipped as Large Turbo wasn't the turbo model at all. 1.3.5 fixes that, and it's built on a much bigger set of automated tests.

The real Large Turbo

What Harvestry downloaded as Large Turbo was OpenAI's full Whisper Large v3: 3.2 GB, and every word it wrote passed through 32 stages. The real Large v3 Turbo does that part in 4, which is where the speed comes from, and it's half the size. 1.3.5 switches to it.

Settings → Tools now shows the real download sizes, too: about 490 MB for Small, 1.5 GB for Medium and 1.6 GB for Large Turbo. Small and Medium had been listed at about half their real size, and the old Large Turbo said ~800 MB but took 3.2 GB.

Timing transcription on its own, with Harvestry's settings, a 58-minute webinar took 29 minutes on our M2 Max before and 5.4 minutes now. About 96% of the words match; most differences are names, medical terms and numbers, and each model gets some right that the other misses. But the old model skipped a 30-second passage, and with the spoken language left on Auto-detect, the new one caught it in all three of our test runs. Screenshot capture runs alongside transcription and hasn't changed, so the whole job won't finish five times sooner.

As for the GPU: Activity Monitor shows the CPU and GPU but has no column for the Neural Engine, so the GPU looks idle and the work seems to be on the CPU. On our Mac, the old model alone kept about two CPU cores busy; the new one keeps about one.

A speed-up we didn't ship

Transcribing the audio in pieces, side by side, made the new model another 1.5 times faster. But it lost six stretches of 15–25 seconds of speech, and some pieces of an English lecture came back in German or Spanish. A transcript that's quietly missing sentences is worse than a slower one.

If you've downloaded Large Turbo

There's nothing to do. The first time you open 1.3.5, Harvestry downloads the new model in the background (about 1.6 GB) and keeps using the old one until it's done, then removes the old 3.2 GB one — or, if the old one is in use at that moment, the next time you open Harvestry. Harvestry now downloads one model at a time, so the other Download buttons in Settings → Tools wait until it's finished. Lectures you've made with Large Turbo won't ask to be processed again. And because the turbo model wasn't trained for translation, choosing Translate to English (Pro) still suggests switching to Medium.

Testing for a bigger Harvestry

Since it launched in June, Harvestry has gained translation, a processing queue, slideshows, PDF documents, Spotlight search and more, and every change can quietly break something somewhere else. That's why much of this release went into automated tests: small programs we can run after every change to check that Harvestry still does what it should.

Harvestry 1.3 had 157 of them; 1.3.5 has 590. Between them, they now put about nine in ten lines of the code that does the work (screenshots, downloads, the processing queue, export, the library) through their paces, up from about half in 1.3. For the windows and buttons themselves, 29 more tests open Harvestry and click through it the way you would, from the welcome screen to processing, Stop Processing and every delete confirmation. They run on a separate virtual Mac with its own throwaway library, so they never touch a real one.

One more test, which we run by hand, puts a real lecture through the real speech model and flags any stretch of 15 seconds or more where someone is talking but nothing was transcribed. On our 58-minute webinar, with the language on Auto-detect, it flags the old Large Turbo's missing 30 seconds, and the new model passes. Writing these tests turned up most of the fixes below.

Things that used to go wrong

Stop Processing (or Abort Processing in the lecture view) now stops consolidation too, the step that writes your consolidated notes, rather than leaving the queue “stopping” for minutes and then saving the notes into the lecture you stopped. If processing stops by itself, for example because a model download was interrupted, the lecture now goes back to Ready instead of staying stuck partway through. A downloaded video can no longer import as one of its temporary pieces instead of the finished file, and Stop Download no longer says “Download failed” or trips up an immediate retry.

Delete All and Delete Unprocessed now also remove the videos Harvestry downloaded from a URL, as deleting a single lecture already did. Videos you imported from your own folders are never touched. Deleted lectures no longer linger in the sidebar, and notes can no longer appear twice in the transcript. And the welcome screen now fits small screens: at 1024×768, Get Started and Don't show this again were off the bottom.

Who gets what

Everything in this release is in both Standard and Pro. Translate to English is part of Pro, so only Pro users (and anyone on the free trial) will see the suggestion to use Medium.

What's in v1.3.5

  • Large Turbo transcribes about 5× faster.In our test, a 58-minute webinar went from 29 minutes to 5.4. It's OpenAI's real turbo model, at half the size.
  • Automatic switch-over.If you've downloaded Large Turbo, the new model arrives in the background, and you get 1.6 GB of disk space back.
  • Real download sizes.Settings → Tools now shows what each model really takes.
  • 590 automated tests.Up from 157, plus 29 that click through the app itself.
  • Stop means stop.Including consolidation.
  • Steadier downloads.No temporary pieces in place of the video, and no false “Download failed.”
  • Cleaner deleting.Delete All and Delete Unprocessed remove the videos Harvestry downloaded from a URL — never your own files.
  • Also fixed.The welcome screen on small screens, screenshots that vanished when iCloud Drive couldn't be written, a newly chosen export folder that sometimes didn't stick, a rare sidebar crash during iCloud sync, and files left behind by a failed ffmpeg install.
v1.3 September 2026

Version 1.3: Now with polish.

1.2 was about new things to do. 1.3 is about how it feels to do them. We went through Harvestry the way a long-time Mac user would — menu by menu, key by key, with VoiceOver on — and fixed everything that didn't feel like it belonged on a Mac. Along the way it picked up one big new trick: any lecture can now become a proper document.

Your lecture, as a document (Pro)

Choose Export → PDF Document… and Harvestry sets the whole lecture as a PDF you'd be happy to print: the transcript in two newspaper-style columns with every slide where it appeared, your highlights and notes exactly as you made them, and the consolidated notes after it — tables and all. It's laid out in the same typeface and warm palette as the study page, at a comfortable reading size on Letter or A4, with page numbers, a clickable link back to the source video, and bookmarks in Preview's sidebar for every section. Slides float to the next column rather than leave gaps, so even a two-hour lecture reads cleanly. Prefer it lighter? PDF Document (Text Only) leaves the slides out.

Find anything

The library has a search field at last — press ⌘F and type to narrow it to matching lectures and folders. And Harvestry now tells Spotlight about your lectures, including everything that was said in them: search your Mac for a phrase you remember from a lecture and the lecture comes up, ready to open. The index is macOS's own and stays on your Mac.

A proper Mac app

There's a new Lecture menu with keyboard shortcuts for everything under the video: Begin Processing (⌘R), Open Study Page (⇧⌘O), Show in Finder (⇧⌘R), Export Markdown (⇧⌘E), Export PDF Document (⌥⌘E), Play/Pause, and Play Slideshow (⌥⌘P). ⌘1, ⌘2 and ⌘3 switch between Transcript, Notes and Slides; the arrow keys move through the library. The window is named after the lecture you're reading, and Harvestry reopens where you left off.

When a lecture finishes processing while you're in another app, the Dock icon counts down the queue and a notification tells you it's ready — click it to jump straight there. There's an Import Video action for the Shortcuts app, too. And deleting a lecture now always asks first: it removes the study page, notes and downloaded video from every Mac, so it should never happen on a stray keypress.

For everyone

Harvestry now works properly with VoiceOver. Every control has a real name, lectures read as their title and status ("Complete, export in iCloud Drive"), folders say how many lectures they hold and whether they're open, and the player, pipeline and screenshots all describe themselves. Things that only appeared on hover are in right-click menus too, and Reduce Motion is respected. Every menu, sheet and message is now translated into all five of Harvestry's languages — and layouts adapt rather than truncate when a German label runs long.

Small things that used to go wrong

Move or rename a video after importing it and Harvestry now follows it, quietly. If it can't — say you copied the video to another drive — the lecture offers Locate Video… to point it at the new file. And if your lectures are all in one language, Settings → Tools now has a Spoken language for new lectures setting, so you don't have to rely on auto-detect. (Both came from requests on our Discord — thank you.)

Who gets what

Everything in this release is in Standard, apart from PDF documents. Search, Spotlight, the Lecture menu and shortcuts, notifications, accessibility, video relinking and the language setting are in both versions. PDF documents join Markdown export and PDF slide handouts in Pro.

What's in v1.3

  • PDF documents (Pro).The whole lecture — transcript, slides, highlights, notes and consolidated notes — as a two-column, printable PDF with bookmarks.
  • Library search.Press ⌘F to find lectures and folders by name.
  • Spotlight.Search your Mac for anything said in a lecture.
  • Lecture menu and keyboard shortcuts.Everything under the video, from the menu bar; ⌘1–3 for the transcript panel's tabs.
  • Notifications and Dock badge.Know when a lecture is ready without watching it.
  • Shortcuts action.Send a video link to Harvestry from the Shortcuts app.
  • VoiceOver and Reduce Motion.Every part of the app, described properly.
  • Fully translated.Every menu, sheet and message in English, German, Spanish, French and Italian.
  • Moved videos, found.Harvestry follows a moved or renamed video, and Locate Video… covers the rest.
  • Default spoken language.Set once in Settings → Tools.
  • Safer deleting.Every delete asks first.
  • Also since 1.2.Drag folders to reorder them (and Sort By now arranges folders too), plus automatic update checks every time Harvestry opens, with a switch in Settings → General.
v1.2 September 2026

Version 1.2: Slides on their own, a queue for whole courses, and sharper screenshots.

Almost everything in this release started as a single list of requests posted on our Discord by someone working through courses one lecture after another. Eight suggestions, all reasonable, most of them things we'd half-meant to do for a while. Together they change how Harvestry feels when you're studying a whole course rather than a single video: line up a week of lectures and walk away, flip through just the slides when that's all you need, and print a handout to take notes on.

Just the slides, when that's what you want

The transcript is the heart of a Harvestry document, but sometimes you don't want to read — you want to see what was on screen. The transcript panel now has a Slides tab showing every captured slide as a grid, each labelled with the moment it appeared. Click one and it opens in a slideshow that fills the window: the arrow keys or a click move from slide to slide, Esc closes it, and one button takes it full screen.

The exported study page has the same thing. A Slides button in the page's controls starts a slideshow from the first slide, and clicking any screenshot now lets you step through all of them with the arrow keys rather than opening one at a time. Because it's part of the page itself, it works for anyone you share the export with — no Harvestry required.

A queue for a whole course

You can now drag videos straight from Finder onto the library sidebar — several at once, or onto a folder to file them there. Each one joins a processing queue and is processed in turn, and the sidebar shows every waiting lecture's place in line. Right-click a lecture to process it next, move it up or down, take it out of the queue, or stop the one that's running. Process All Unprocessed, in the library's menu, queues everything you haven't processed yet in one go.

The queue runs one lecture at a time, and that's deliberate. Each lecture already keeps the Neural Engine and the video decoder fully busy, so running two side by side doesn't finish either one sooner — it just makes both slower and makes your Mac work harder. Starting a second lecture while one was still processing used to do exactly that; now it simply waits its turn.

Sharper slides

Screenshots are now captured at the video's own resolution, up to 4K. Until now they topped out at 2560×1440, which mattered for anyone importing 4K recordings of dense slides.

The bigger change for online video is a new Download quality setting under Settings → Tools. It stays at 1080p by default: those downloads are quick and ready to process straight away. Choose Best available and Harvestry takes the highest resolution the site offers — up to 4K on YouTube — which gives noticeably crisper slides when the text is small. The honest trade-off: above 1080p, most sites only offer formats that have to be converted before processing, so downloads are larger and take longer, and each lecture uses more iCloud storage. For most lectures, 1080p is plenty.

PDF handouts (Pro)

Harvestry Pro can now turn a lecture's slides into a printable PDF, straight from the Slides tab, in three layouts: one large slide per page for reading on screen, three per page beside ruled lines for taking notes by hand, or six per page as a compact overview. Pages are Letter or A4 depending on where you are, and every slide carries its timestamp so you can find the moment again in the recording.

Who gets what

Everything in this release is in Standard, apart from PDF handouts. The Slides tab, both slideshows, the processing queue, drag-and-drop, sorting, 4K capture, and the new download setting are all included in both versions. PDF handouts join Markdown export as part of Pro — a new kind of export alongside the HTML page every edition already has.

What's in v1.2

  • Slides tab and slideshow.See every captured slide as a grid and flip through them full-window with the arrow keys, a click, or full screen.
  • Slideshow in the exported page.A Slides button on the study page, and arrow-key navigation between screenshots — for you and anyone you share it with.
  • Processing queue.Line up as many lectures as you like; they process one after another, and you can reorder, remove, or stop them from the sidebar.
  • Drag and drop into the library.Drop videos from Finder onto the sidebar — or onto a folder — and they're imported and queued. Files already in your library are skipped.
  • Sort by title.Keep the library in date order, or switch to alphabetical under the library's menu. Numbers sort naturally, so Lecture 2 comes before Lecture 10.
  • Screenshots up to 4K.Slides are captured at the video's native resolution, up to 4K.
  • Download quality setting.Keep the fast 1080p default, or download the best quality a site offers for sharper slides.
  • PDF handouts (Pro).One, three with note lines, or six slides per page, ready to print.
  • Fixes since 1.1.Including a native Apple Silicon build of ffmpeg, which fixes URL imports and video conversion on Macs without Rosetta, and an in-app yt-dlp update that now actually finishes — it could previously stall and time out, leaving the old version in place.
v1.1 September 2026

Version 1.1: Translation, plus ChatGPT consolidation.

Since 1.0 arrived in June there have been six releases — bug fixes and small feature updates, most of them shaped by people writing in to tell us what was getting in their way. The reception has been warmer than we had any right to expect, and we're genuinely grateful for it. Version 1.1 is the first release with two larger additions: translating foreign-language audio into English, and ChatGPT as a new option for the consolidation step.

Transcribe, translate, and the difference between them

Harvestry has always transcribed — turning speech into text in the same language it was spoken in. A French lecture produced a French transcript. That's still what happens by default, and it's included in both Standard and Pro.

What's new is translate: a French lecture, a Japanese conference talk, a German seminar, rendered directly into an English transcript in a single pass. It isn't a second step bolted onto the end — the model goes from foreign-language audio to English text itself, on your Mac, with no cloud service and no audio leaving the machine. Everything downstream behaves normally: screenshots, timestamps, audio sync, highlights, and consolidation all work on the English result.

Two honest caveats before you buy. The first is that quality is not uniform across the languages Whisper supports. French, Spanish, German, Italian, Portuguese, Dutch and other widely spoken languages translate very well. Low-resource languages — Maltese, Amharic, Shona, Lao, Yoruba and similar — are supported, but the results are markedly weaker, and the failure mode is fluent-sounding English that doesn't match what was actually said rather than obvious nonsense. If your material is in one of those, test it during the trial before committing. Everything runs locally and costs nothing to re-run.

The second is that translation runs into English. That's a constraint of the speech model itself, not a choice we made — it was trained to translate into English and no other target language. French to English works beautifully. French to German is not something the transcription stage can do.

But that pairing isn't out of reach — it just happens one step later. Version 1.1 also adds a Notes language setting for consolidation (Pro). Transcribe a French lecture in French, set Notes language to German, and the consolidation model writes your study notes in German. Because it works from the text rather than the audio, it isn't bound to English, and for less common languages it tends to do better than speech translation does. You'll find it under Settings → Consolidation, and the documentation walks through an example.

Alongside this, you can now tell Harvestry what language is being spoken instead of letting it guess. Detection samples only the opening of the audio, so a video that starts with a music bed, a silent title card, or an English introduction over a French programme can send it down the wrong path — and once it commits, the whole transcript follows. Setting the language explicitly takes one click and removes the guesswork. Both options live in the same menu on the Transcription step, and the export is regenerated if you change your mind after processing.

Consolidation, now with ChatGPT

Consolidation is the optional step that takes your raw, verbatim transcript and asks a language model to turn it into organised notes — headings, structure, the argument pulled out of the digressions. You get both documents in the export: the faithful record and the refined version, side by side.

Until now that meant the Claude API, or a local Ollama model if you have Pro. Version 1.1 adds OpenAI's ChatGPT as a third option. Paste your own API key, pick a model, and it works exactly like the others — the model list is fetched live from OpenAI, so new releases appear in the picker without waiting for a Harvestry update. Switching providers on an already-processed lecture offers to regenerate the notes and re-export, so you can compare how different models handle the same material.

As before, this is the one step that leaves your Mac, it only runs if you turn it on, and it uses your own API key and your own account. If you'd rather nothing left the machine at all, local Ollama consolidation remains the Pro answer.

Who gets what

Transcription is in both versions. Speech to text in the language it was spoken — English to English, French to French, and every other language the model supports — is part of Standard and always has been. Choosing the spoken language explicitly is included too.

Translation to English is a Pro feature. If you're on Standard you'll see the option with a lock beside it, and nothing changes about the transcription you already rely on.

What's in v1.1

  • Translate to English (Pro).Turn foreign-language audio directly into an English transcript in one on-device pass — no cloud service, no audio leaving your Mac, and every downstream feature working as usual.
  • ChatGPT consolidation.Use OpenAI models for the consolidation step with your own API key, alongside the existing Claude API and local Ollama options. The model list is fetched live, so new releases show up on their own.
  • Notes in any language (Pro).Have the consolidation model write your study notes in a different language from the transcript — French to German, Japanese to Spanish, any pair the model knows. Set it under Settings → Consolidation → Notes language.
  • Spoken-language selection.Pick the source language yourself instead of relying on auto-detection, which only samples the opening of the audio and can be thrown by intros, music, or silence. Included in both versions.
  • Regenerate after changing your mind.Changing the language, the translation setting, the transcription model, or the consolidation provider on a finished lecture now offers to reprocess and re-export, keeping your existing notes and annotations intact.
  • Six releases of fixes since 1.0.Faster library loading, more reliable iCloud sync, steadier exports, and a long tail of smaller corrections — nearly all of them reported by people who took the time to write in.
v1.0 June 2026

Harvestry is here.

We built Harvestry because rewatching a lecture to find one slide is a terrible use of an hour. Version 1.0 is now available — it turns any lecture video into a polished, searchable study document in minutes, entirely on your Mac.

Import a video or paste a URL, press Begin Processing, and Harvestry handles the rest. Transcription runs on Apple's Neural Engine while screenshot capture works through the video track simultaneously — by the time both finish, your document is ready. No cloud. No account. No waiting on a server.

The output is a self-contained HTML file: warm editorial typography, live audio sync, per-word highlighting, dark mode, and a font-size slider — all built in and dependency-free. It opens instantly in any browser and works offline forever.

What's in v1.0

  • On-device transcription.Five Whisper model sizes — Tiny through Large Turbo — running entirely on Apple Silicon. No audio ever leaves your Mac.
  • Intelligent screenshot capture.Scene-change detection targets genuine content transitions, not presenter movement. A sharp face in the corner won't rescue a blurry slide — the frame is rejected and the next clear one is used instead.
  • Manual curation.Full scrubber timeline, Seek to Clear (jumps to the nearest sharp frame in either direction), Add to Transcript (captures any frame on demand), inline notes, and title image selection.
  • Annotations baked into the export.Highlights in four colours, margin notes connected to underlined phrases by a dotted line, and inline notes between passages — all embedded in the HTML and travelling with the document.
  • URL import.Paste a YouTube, Vimeo, or any of 1,500+ platform URLs — including login-protected and university-hosted videos — and Harvestry downloads and processes it directly.
  • Five languages.English, German, Italian, Spanish, and French — fully localised interface and export output, with a language picker right on the welcome screen.
  • LLM consolidation.Optionally send your transcript to the Claude API to produce a refined, consolidated set of notes alongside the verbatim record. Both versions export together.
  • Local Ollama consolidation (Pro).Run the same consolidation entirely on-device using any locally installed Ollama model. No data leaves your Mac, no per-token cost, no internet required.
  • Obsidian & Markdown export (Pro).Export as a Markdown file with YAML frontmatter, linked screenshot assets, and a per-lecture folder structure — ready to drop straight into your vault or any Markdown editor.
  • iCloud library sync.Your full collection of transcripts, screenshots, and study documents syncs automatically across up to two Macs via iCloud Drive.
  • Privacy by design.No account, no telemetry, no analytics. Harvestry processes your content locally and keeps its business to itself.